[RULE] Spreadsheet/DB of the SW, part 2

[M. Fioretti]

On Wed, Feb 26, 2003 09:42:18 at 09:42:18AM -0800, Eugene Wong  disposable_eugene at hotmail.com  wrote:
> 
> 1 of my concerns with that web page is that it may take a lot of 
> calculations each time someone queries the database. 
That depends from how it is done, ie from how much meaningful
information is already precalculated and stored inside it. We have to
experiment on this. It would also be possible to make available a tar
file with:
	the database snapshot, to recreate it locally
	a script to query it on your machine, no server
	overload/bandwidth bottleneck

> I misunderstand you? With a bash script, you can do something like, 
> "./install_script base_list my_file_list", & then have it install the base 
> rpms 1st, then the files of your choice.
No, that is just what I meant (we can leave details on the syntax and
formatting of my_file_list to another moment), with my_file_list
generated (maybe days) *before* the install, by qyerying the online
database, and then saved somewhere

> 
> >7) Why should this script/spreadsheet thingy happen just during the
> >install, or be limited by busybox/floppy space constraints? In many
> >cases, I want to know and tune the possible package list (days?)
> >*before* installation (maybe working on it ten minutes at a time in
> >the office during lunch breaks, see URL above again), so that, when I
> >actually start the install, I can walk away or take a nap.
> 
> I agree. I would like to see a script where the user can go browse through 
> all of the rpms available, & choose as you describe.
> 
> >In the second place, the script using the db/spreadsheet can be run
> >automatically *after* the install: you install the very base packages,
> >then (automatically) reboot and run that script with full availability
> >of wget, Perl, ssh, whatever. As long as it can be automated, and
> >takes almost the same total time, what difference does it make?
> 
> I am very much in favour of rebooting. This type of a philosophy allows us 
> to upgrade much more easily, without having to re-enter information. Right 
> now, I am trying to create some install scripts that can be used long after 
> the install.
> 
Looks like we are all converging on this, doesn't it? (see also
Colin's thread)

> I would like to go back to discussing the database.
> 
> Perhaps what I am really looking for is a general menu of functionality, 
I love this. The database, or whatever else, should ask: do you need
IMAP email (not do you want Kmail or mutt) Do you need Javascript
support (not : do you need mozilla or links) I've been saying this for
years. 
-- 
Marco Fioretti                 mfioretti
Red Hat for low memory         www.rule-project.org

We are drowning in information but starved for knowledge.
                                      -- John Naisbitt, Megatrends


_______________________________________________
Original home page of the RULE project: www.rule-project.org
Original Rule Development Site http://savannah.gnu.org/projects/rule/
Original RULE mailing list: Rule-list at nongnu.org, hosted at http://mail.nongnu.org/mailman/listinfo/rule-list